Further Statistics: Discrete Distributions Depth - Worksheets, Questions and Revision
21 original exam-style questions - 7 pages of questions with a full mark scheme - free printable PDF.
Further Statistics: Discrete Distributions Depth revision guideWrite-in editionMore A Level Further Maths practiceMock exams
This topic is chapter 5 of A Level Further Maths Further Statistics Topic Workbook and chapter 15 of A Level Further Maths 10-Minute Tests 2.
2.5 Further Statistics: Discrete Distributions Depth
Key facts: discrete distributions and probability generating functions
Original content written for Revision Library.
For X ~ Po(lambda), P(X = x) = e^(-lambda) * lambda^x / x!, for x = 0, 1, 2, ..., with E(X) = lambda and Var(X) = lambda. If X ~ Po(lambda1) and Y ~ Po(lambda2) are independent, then X + Y ~ Po(lambda1 + lambda2).
For X ~ Geo(p), the number of independent trials up to and including the first success (write q = 1 - p), P(X = x) = q^(x-1) * p, for x = 1, 2, 3, ..., with E(X) = 1/p and Var(X) = q/p^2. The geometric distribution has the memoryless property: P(X > m+n | X > m) = P(X > n) for positive integers m and n.
For X ~ NB(r, p), the number of independent trials up to and including the rth success, P(X = x) = C(x-1, r-1) * p^r * q^(x-r), for x = r, r+1, r+2, ..., with E(X) = r/p and Var(X) = rq/p^2. X can be written as the sum of r independent Geo(p) random variables.
The probability generating function of a discrete random variable X taking non-negative integer values is G_X(t) = E(t^X) = sum of P(X = x) * t^x. It satisfies G_X(1) = 1, E(X) = G_X'(1), and Var(X) = G_X''(1) + G_X'(1) - [G_X'(1)]^2. If X and Y are independent random variables, then G_(X+Y)(t) = G_X(t) * G_Y(t).
| Y | 1 | 2 | 3 | 4 |
|---|---|---|---|---|
| P(Y = y) | 0.1 | 0.4 | 0.3 | 0.2 |
| X | 0 | 1 | 2 |
|---|---|---|---|
| P(X = x) | 0.2 | 0.3 | 0.5 |
Question 1
- (a) correct use of P(X=5) = q4 * p with q = 0.65, p = 0.35, oeM1
- (a) awrt 0.0625A1
- (a) Answer: P(X = 5) = awrt 0.0625
- (b) recognises P(X>n) = qn, i.e. the probability the first 3 attempts all failM1
- (b) 0.274625, awrt 0.275A1
- (b) Answer: P(X > 3) = 0.274625, awrt 0.275
- (c) E(X) = 1/0.35 = 20/7, awrt 2.86, with a valid interpretation in context, oeB1
- (c) Answer: E(X) = 20/7 = awrt 2.86; on average, Meera needs about 2.86 attempts up to and including her first success.
Question 2
- (a) uses E(X) = 1/p = 12.5, oeM1
- (a) p = 0.08, caoA1
- (a) Answer: p = 0.08
- (b) uses Var(X) = (1-p)/p2 = 20, oeM1
- (b) rearranges to 1 - p = 20p2M1
- (b) 20p2 + p - 1 = 0, csoA1
- (b) solves the quadratic, e.g. p = (-1 ± √1+80)/40 = (-1 ± 9)/40M1
- (b) p = 0.2 (rejecting p = -0.25 as 0 < p < 1), caoA1
- (b) Answer: p = 0.2
Question 3
- (a) GY(t) = sum of P(Y=y)*ty, using the tableM1
- (a) GY(t) = 0.1t + 0.4t2 + 0.3t3 + 0.2t4, oeA1
- (a) Answer: GY(t) = 0.1t + 0.4t2 + 0.3t3 + 0.2t4
- (b) differentiates: GY'(t) = 0.1 + 0.8t + 0.9t2 + 0.8t3M1
- (b) E(Y) = GY'(1) = 2.6, caoA1
- (b) differentiates again: GY''(t) = 0.8 + 1.8t + 2.4t2, so GY''(1) = 5.0M1
- (b) uses Var(Y) = GY''(1) + GY'(1) - [GY'(1)]2, oeM1
- (b) Var(Y) = 0.84, caoA1
- (b) Answer: E(Y) = 2.6, Var(Y) = 0.84
- (c) uses E(Y) = sum of y*P(Y=y) = 1(0.1)+2(0.4)+3(0.3)+4(0.2)M1
- (c) = 2.6, matching part (b), caoA1
- (c) Answer: E(Y) = 0.1+0.8+0.9+0.8 = 2.6, matching part (b).
Question 4
- (a) n is large and p is smallB1
- (a) X is approximately Po(100 x 0.02) = Po(2)B1
- (a) Answer: Po(2)
- (b) 100C2 x 0.022 x 0.9898M1
- (b) 0.2734A1
- (b) e-2 22/2!M1
- (b) 0.2707A1
- (b) Answer: binomial 0.2734, Poisson 0.2707
- (c) |approximate - exact|/exact x 100M1
- (c) 1.0%A1
- (c) Answer: 1.0%
Question 5
- (a) uses the sum of all probabilities = 1M1
- (a) forms the sum P(X = 1) + P(X = 2) + P(X = 3) + P(X = 4) in terms of kM1
- (a) k(30) = 1, so k = 1/30 (cso)A1
- (a) Answer: k = 1/30
- (b) E(X) = sum of x P(X = x)M1
- (b) E(X) = 10/3A1
- (b) Var(X) = E(X2) - [E(X)]2M1
- (b) E(X2) = 59/5, Var(X) = 31/45A1
- (b) Answer: E(X) = 10/3, Var(X) = 31/45
- (c) E(aX + b) = aE(X) + bM1
- (c) 38/3A1
- (c) Var(aX + b) = a2 Var(X)M1
- (c) 155/9A1
- (c) Answer: E = 38/3, Var = 155/9
Question 6
- (a) sum of P(X = i)P(Y = i) over i = 1, 2, 3, using independenceM1
- (a) 9/20A1
- (a) Answer: 9/20
- (b) adds P(X = i)P(Y = j) over the cases i > jM1
- (b) 13/40A1
- (b) Answer: 13/40
- (c) E(X + Y) = E(X) + E(Y)M1
- (c) 31/8A1
- (c) Var(X - Y) = Var(X) + Var(Y) for independent variablesM1
- (c) 243/320A1
- (c) Answer: E(X + Y) = 31/8, Var(X - Y) = 243/320
Question 7
- (a) P(X = 5) = e-2 25/5!M1
- (a) 0.0361A1
- (a) P(X ≥ 5) = 1 - P(X ≤ 4)M1
- (a) 0.0527A1
- (a) Answer: 0.0361; 0.0527
- (b) E(X) = Var(X) = 2B1
- (b) for a Poisson distribution the variance equals the meanB1
- (b) Answer: E(X) = Var(X) = 2
- (c) X + Y ~ Po(2 + 1.5) = Po(3.5)M1
- (c) P = e-3.5 3.55/5!A1
- (c) 0.132A1
- (c) Answer: 0.132
Question 8
- (a) sum of P(X = i)P(Y = i) over i = 1, 2, 3, using independenceM1
- (a) 17/49A1
- (a) Answer: 17/49
- (b) adds P(X = i)P(Y = j) over the cases i > jM1
- (b) 27/98A1
- (b) Answer: 27/98
- (c) E(X + Y) = E(X) + E(Y)M1
- (c) 29/7A1
- (c) Var(X - Y) = Var(X) + Var(Y) for independent variablesM1
- (c) 55/49A1
- (c) Answer: E(X + Y) = 29/7, Var(X - Y) = 55/49
Question 9
- (a) P(X = 6) = (1 - 0.25)5 x 0.25M1
- (a) 0.0593A1
- (a) P(X > 6) = (1 - 0.25)6M1
- (a) 0.178A1
- (a) Answer: 0.0593; 0.178
- (b) E(X) = 1/pM1
- (b) 4A1
- (b) Var(X) = (1 - p)/p2M1
- (b) 12.00A1
- (b) Answer: E(X) = 4.000, Var(X) = 12.00
- (c) P(X ≤ n) = 1 - (1 - 0.25)n > 0.99M1
- (c) (1 - 0.25)n < 0.01, and takes logs (changing the inequality)M1
- (c) n > 16.008A1
- (c) n = 17A1
- (c) Answer: n = 17
Question 10
- (a) sum of P(X = i)P(Y = i) over i = 1, 2, 3, using independenceM1
- (a) 49/135A1
- (a) Answer: 49/135
- (b) adds P(X = i)P(Y = j) over the cases i > jM1
- (b) 64/135A1
- (b) Answer: 64/135
- (c) E(X + Y) = E(X) + E(Y)M1
- (c) 208/45A1
- (c) Var(X - Y) = Var(X) + Var(Y) for independent variablesM1
- (c) 2426/2025A1
- (c) Answer: E(X + Y) = 208/45, Var(X - Y) = 2426/2025
Question 11
- (a) sum of P(X = i)P(Y = i) over i = 1, 2, 3, using independenceM1
- (a) 1/3A1
- (a) Answer: 1/3
- (b) adds P(X = i)P(Y = j) over the cases i > jM1
- (b) 11/26A1
- (b) Answer: 11/26
- (c) E(X + Y) = E(X) + E(Y)M1
- (c) 551/156A1
- (c) Var(X - Y) = Var(X) + Var(Y) for independent variablesM1
- (c) 27587/24336A1
- (c) Answer: E(X + Y) = 551/156, Var(X - Y) = 27587/24336
Question 12
- (a) P(X = 3) = e-3.5 3.53/3!M1
- (a) 0.216A1
- (a) P(X ≥ 3) = 1 - P(X ≤ 2)M1
- (a) 0.679A1
- (a) Answer: 0.216; 0.679
- (b) E(X) = Var(X) = 3.5B1
- (b) for a Poisson distribution the variance equals the meanB1
- (b) Answer: E(X) = Var(X) = 3.5
- (c) X + Y ~ Po(3.5 + 1) = Po(4.5)M1
- (c) P = e-4.5 4.53/3!A1
- (c) 0.169A1
- (c) Answer: 0.169
Question 13
- (a) uses the sum of all probabilities = 1M1
- (a) forms the sum P(X = 1) + P(X = 2) + P(X = 3) + P(X = 4) in terms of kM1
- (a) k(30) = 1, so k = 1/30 (cso)A1
- (a) Answer: k = 1/30
- (b) E(X) = sum of x P(X = x)M1
- (b) E(X) = 10/3A1
- (b) Var(X) = E(X2) - [E(X)]2M1
- (b) E(X2) = 59/5, Var(X) = 31/45A1
- (b) Answer: E(X) = 10/3, Var(X) = 31/45
- (c) E(aX + b) = aE(X) + bM1
- (c) 8/3A1
- (c) Var(aX + b) = a2 Var(X)M1
- (c) 124/45A1
- (c) Answer: E = 8/3, Var = 124/45
Question 14
- (a) P(X = 2) = e-5 52/2!M1
- (a) 0.0842A1
- (a) P(X ≥ 2) = 1 - P(X ≤ 1)M1
- (a) 0.960A1
- (a) Answer: 0.0842; 0.960
- (b) E(X) = Var(X) = 5B1
- (b) for a Poisson distribution the variance equals the meanB1
- (b) Answer: E(X) = Var(X) = 5
- (c) X + Y ~ Po(5 + 2) = Po(7)M1
- (c) P = e-7 72/2!A1
- (c) 0.0223A1
- (c) Answer: 0.0223
Question 15
- (a) E(X) = r/p = 3/0.08 = 37.5, caoB1
- (a) Answer: E(X) = 37.5
- (b) identifies x = 10, r = 3 and uses P(X=x) = C(x-1,r-1) * pr * qx-r, oeM1
- (b) correct unsimplified expression C(9,2) * (0.08)3 * (0.92)7, oeA1
- (b) awrt 0.0103A1
- (b) Answer: P(X = 10) = awrt 0.0103
- (c) Var(X) = r*q/p2 = 3*0.92/0.0064 = 431.25, caoB1
- (c) Answer: Var(X) = 431.25
Question 16
- (a) GX(t) = sum of P(X=x) * tx, using the probabilities givenM1
- (a) GX(t) = 0.2 + 0.3t + 0.5t2, oeA1
- (a) Answer: GX(t) = 0.2 + 0.3t + 0.5t2
- (b) GX(1) = 0.2 + 0.3 + 0.5 = 1B1
- (b) valid explanation, e.g. GX(1) = sum of all the probabilities in the distribution, which must total 1 for any valid probability distribution, oeB1
- (b) Answer: GX(1) = 1, since the probabilities in any distribution must sum to 1.
- (c) differentiates: GX'(t) = 0.3 + t, oeM1
- (c) E(X) = GX'(1) = 1.3, caoA1
- (c) Answer: E(X) = 1.3
Question 17
- (a) writes GX(t) = sum from x=0 to infinity of tx * e-λ * λx / x!, oeM1
- (a) factors out e-λ and combines terms: GX(t) = e-λ * sum from x=0 to infinity of (λ*t)x / x!M1
- (a) recognises sum from x=0 to infinity of (λ*t)x/x! = eλ*t (the exponential series)M1
- (a) GX(t) = e-λ * eλ*t = eλ(t-1), csoA1
- (a) Answer: GX(t) = eλ(t-1)
- (b) differentiates: GX'(t) = λ * eλ(t-1)M1
- (b) substitutes t = 1 into GX'(t)dM1
- (b) E(X) = GX'(1) = λ * e0 = λ, csoA1
- (b) Answer: E(X) = λ
- (c) differentiates again: GX''(t) = λ2 * eλ(t-1), so GX''(1) = λ2M1
- (c) uses Var(X) = GX''(1) + GX'(1) - [GX'(1)]2 = λ2 + λ - λ2, oeM1
- (c) Var(X) = λ, csoA1
- (c) Answer: Var(X) = λ
Question 18
- (a) GX(t) = elambda1(t-1) and GY(t) = elambda2(t-1), both stated (ft from Q17, or quoted)B1
- (a) Answer: GX(t) = elambda1(t-1), GY(t) = elambda2(t-1)
- (b) writes GW(t) = E(tX+Y) = E(tX * tY), oeM1
- (b) uses independence of X and Y: E(tX * tY) = E(tX)*E(tY) = GX(t)*GY(t)M1
- (b) combines: GW(t) = elambda1(t-1) * elambda2(t-1) = e(lambda1+lambda2)(t-1)M1
- (b) identifies this as the PGF of a Po(lambda1+lambda2) distribution; since a PGF uniquely determines a distribution, W ~ Po(lambda1+lambda2), csoA1
- (b) Answer: W ~ Po(lambda1 + lambda2)
- (c) identifies W ~ Po(2.4+3.1) = Po(5.5), using the result of part (b), oeM1
- (c) correct use of P(W=5) = e-5.5 * 5.55 / 5!M1
- (c) awrt 0.171A1
- (c) Answer: P(W = 5) = awrt 0.171
Question 19
- (a) states or derives P(X > k) = qk for a positive integer k, oeB1
- (a) uses P(X>m+n | X>m) = P(X>m+n and X>m)/P(X>m) = P(X>m+n)/P(X>m), since {X>m+n} is a subset of {X>m}M1
- (a) substitutes: qm+n/qmM1
- (a) = qn = P(X>n), as required, csoA1
- (a) Answer: P(X > m+n | X > m) = qn = P(X > n) (shown)
- (b) uses the memoryless property from part (a): required probability = P(X>15) = q15, oeM1
- (b) awrt 0.542A1
- (b) Answer: 0.542 (3 s.f.)
Question 20
- (a) writes GX(t) = sum from x=1 to infinity of tx * qx-1 * p, oeM1
- (a) takes out a factor of pt: GX(t) = pt * sum from k=0 to infinity of (qt)k, oeM1
- (a) recognises sum from k=0 to infinity of (qt)k = 1/(1-qt) as an infinite geometric series, valid for |qt| < 1M1
- (a) GX(t) = pt/(1-qt), csoA1
- (a) Answer: GX(t) = pt/(1-qt)
- (b) differentiates using the quotient rule: GX'(t) = p*(1-qt)-1 + p*q*t*(1-qt)-2, oeM1
- (b) substitutes t = 1, using 1 - q = pdM1
- (b) E(X) = GX'(1) = p/p + pq/p2 = 1 + q/p = (p+q)/p = 1/p, csoA1
- (b) Answer: E(X) = 1/p
- (c) differentiates GX'(t) again to obtain GX''(t) = 2pq*(1-qt)-2 + 2pq2*t*(1-qt)-3, oeM1
- (c) substitutes t = 1 to get GX''(1) = 2q/p + 2q2/p2dM1
- (c) uses Var(X) = GX''(1) + GX'(1) - [GX'(1)]2 and combines all terms over a common denominator p2dM1
- (c) Var(X) = q/p2, cso, using p+q=1 to simplify the numerator to qA1
- (c) Answer: Var(X) = q/p2
Question 21
- (a) GXi(t) = pt/(1-qt), ft from Q20(a) or quotedB1
- (a) Answer: GXi(t) = pt/(1-qt)
- (b) states X is the sum of r independent, identically distributed random variablesM1
- (b) uses the PGF-of-a-sum result (extended to r independent variables): the PGF of a sum of independent variables is the product of their individual PGFsM1
- (b) GX(t) = [pt/(1-qt)]r, csoA1
- (b) Answer: GX(t) = [pt/(1-qt)]r
- (c) lets g(t) = pt/(1-qt), so GX(t) = [g(t)]r; differentiates using the chain rule: GX'(t) = r*[g(t)]r-1 * g'(t)M1
- (c) uses g(1) = 1 and g'(1) = 1/p (from Q6)M1
- (c) substitutes t = 1: GX'(1) = r * 1r-1 * (1/p)dM1
- (c) E(X) = GX'(1) = r/p, csoA1
- (c) Answer: E(X) = r/p
- (d) differentiates again using the product/chain rule: GX''(t) = r(r-1)*[g(t)]r-2*[g'(t)]2 + r*[g(t)]r-1*g''(t)M1
- (d) substitutes t=1 using g(1)=1, g'(1)=1/p, g''(1)=2q/p+2q2/p2 (from Q6) to get GX''(1) = r(r-1)/p2 + 2rq/p + 2rq2/p2M1
- (d) uses Var(X) = GX''(1) + GX'(1) - [GX'(1)]2 and combines all terms over a common denominator p2dM1
- (d) Var(X) = rq/p2, cso, using p+q=1 to simplify the numerator to rqA1
- (d) Answer: Var(X) = rq/p2
Mark your answers
Check answers without an account. Some answers can be checked automatically; use the mark scheme to mark the rest.